arXiv subjects

Dong Liu

Publications and source records attributed to Dong Liu.

At least 91 records · Page 5Linked to original sources

arXiv2025

Distributed Reinforcement Learning using Local Smart Meter Data for Voltage Regulation in Distribution Networks

Centralised reinforcement learning (RL) for voltage magnitude regulation in distribution networks typically involves numerous agent-environment interactions and power flow (PF) calculations, inducing computational overhead and privacy concerns over shared data. Thus, we propose a distributed RL algorithm to regulate voltage magnitude. First, a dynamic Thevenin equivalent model is integrated within smart meters (SM), enabling local voltage magnitude estimation using local SM data for RL agent training, and mitigating the dependency of synchronised data collection and centralised PF calculations. To mitigate estimation errors induced by Thevenin model inaccuracies, a voltage magnitude correction strategy that combines piecewise functions and neural networks is introduced. The piecewise function corrects the large errors of estimated voltage magnitude, while a neural network mimics the grid's sensitivity to control actions, improving action adjustment precision. Second, a coordination strategy is proposed to refine local RL agent actions online, preventing voltage magnitude violations induced by excessive actions from multiple independently trained agents. Case studies on energy storage systems validate the feasibility and effectiveness of the proposed approach, demonstrating its potential to improve voltage regulation in distribution networks.

eess.SY↗

arXiv2025

CXL-SpecKV: A Disaggregated FPGA Speculative KV-Cache for Datacenter LLM Serving

Large Language Models (LLMs) have revolutionized natural language processing tasks, but their deployment in datacenter environments faces significant challenges due to the massive memory requirements of key-value (KV) caches. During the autoregressive decoding process, KV caches consume substantial GPU memory, limiting batch sizes and overall system throughput. To address these challenges, we propose \textbf{CXL-SpecKV}, a novel disaggregated KV-cache architecture that leverages Compute Express Link (CXL) interconnects and FPGA accelerators to enable efficient speculative execution and memory disaggregation. Our approach introduces three key innovations: (i) a CXL-based memory disaggregation framework that offloads KV-caches to remote FPGA memory with low latency, (ii) a speculative KV-cache prefetching mechanism that predicts and preloads future tokens' cache entries, and (iii) an FPGA-accelerated KV-cache compression and decompression engine that reduces memory bandwidth requirements by up to 4$\times$. When evaluated on state-of-the-art LLM models, CXL-SpecKV achieves up to 3.2$\times$ higher throughput compared to GPU-only baselines, while reducing memory costs by 2.8$\times$ and maintaining accuracy. Our system demonstrates that intelligent memory disaggregation combined with speculative execution can effectively address the memory wall challenge in large-scale LLM serving. Our code implementation has been open-sourced at https://github.com/FastLM/CXL-SpecKV.

cs.AI↗

arXiv2025

MM-ACT: Learn from Multimodal Parallel Generation to Act

A generalist robotic policy needs both semantic understanding for task planning and the ability to interact with the environment through predictive capabilities. To tackle this, we present MM-ACT, a unified Vision-Language-Action (VLA) model that integrates text, image, and action in shared token space and performs generation across all three modalities. MM-ACT adopts a re-mask parallel decoding strategy for text and image generation, and employs a one-step parallel decoding strategy for action generation to improve efficiency. We introduce Context-Shared Multimodal Learning, a unified training paradigm that supervises generation in all three modalities from a shared context, enhancing action generation through cross-modal learning. Experiments were conducted on the LIBERO simulation and Franka real-robot setups as well as RoboTwin2.0 to assess in-domain and out-of-domain performances respectively. Our approach achieves a success rate of 96.3% on LIBERO, 72.0% across three tasks of real Franka, and 52.38% across eight bimanual tasks of RoboTwin2.0 with an additional gain of 9.25% from cross-modal learning. We release our codes, models and data at https://github.com/HHYHRHY/MM-ACT.

cs.CV↗

arXiv2025

Formally and Empirically Verified Methodologies for Scalable Hierarchical Full-Stack Systems

This paper introduces Primary Breadth-First Development (PBFD) and Primary Depth-First Development (PDFD)-formally and empirically verified methodologies for scalable, industrial-grade full-stack software engineering. Both approaches enforce structural and behavioral correctness through graph-theoretic modeling, bridging formal methods and real-world practice. PBFD and PDFD model software development as layered directed graphs with unified state machines, verified using Communicating Sequential Processes (CSP) and Linear Temporal Logic (LTL). This guarantees bounded-refinement termination, deadlock freedom, and structural completeness. To manage hierarchical data at scale, we present the Three-Level Encapsulation (TLE)-a novel bitmask-based encoding scheme. TLE operations are verified via CSP failures-divergences refinement, ensuring constant-time updates and compact storage that underpin PBFD's robust performance. PBFD demonstrates exceptional industrial viability through eight years of enterprise deployment with zero critical failures, achieving approximately 20x faster develop-ment than Salesforce OmniScript, 7-8x faster query performance, and 11.7x storage reduction compared to conventional relational models. These results are established through longitudinal observational studies, quasi-experimental runtime comparisons, and controlled schema-level experiments. Open-source Minimum Viable Product implementations validate key behavioral properties, including bounded refinement and constant-time bitmask operations, un-der reproducible conditions. All implementations, formal specifications, and non-proprietary datasets are publicly available.

cs.SE↗

arXiv2025

Fiber Bundle Networks: A Geometric Machine Learning Paradigm

We propose Fiber Bundle Networks (FiberNet), a novel machine learning framework integrating differential geometry with machine learning. Unlike traditional deep neural networks relying on black-box function fitting, we reformulate classification as interpretable geometric optimization on fiber bundles, where categories form the base space and wavelet-transformed features lie in the fibers above each category. We introduce two innovations: (1) learnable Riemannian metrics identifying important frequency feature components, (2) variational prototype optimization through energy function minimization. Classification is performed via Voronoi tessellation under the learned Riemannian metric, where each prototype defines a decision region and test samples are assigned to the nearest prototype, providing clear geometric interpretability. This work demonstrates that the integration of fiber bundle with machine learning provides interpretability and efficiency, which are difficult to obtain simultaneously in conventional deep learning.

cs.LG↗

arXiv2025

From Betti Numbers to Persistence Diagrams: A Hybrid Quantum Algorithm for Topological Data Analysis

Persistence diagrams serve as a core tool in topological data analysis, playing a crucial role in pathological monitoring, drug discovery, and materials design. However, existing quantum topological algorithms, such as the LGZ algorithm, can only efficiently compute summary statistics like Betti numbers, failing to provide persistence diagram information that tracks the lifecycle of individual topological features, severely limiting their practical value. This paper proposes a novel quantum-classical hybrid algorithm that achieves, for the first time, the leap from "quantum computation of Betti numbers" to "quantum acquisition of practical persistence diagrams." The algorithm leverages the LGZ quantum algorithm as an efficient feature extractor, mining the harmonic form eigenvectors of the combinatorial Laplacian as well as Betti numbers, constructing specialized topological kernel functions to train a quantum support vector machine (QSVM), and learning the mapping from quantum topological features to persistence diagrams. The core contributions of this algorithm are: (1) elevating quantum topological computation from statistical summaries to pattern recognition, greatly expanding its application value; (2) obtaining more practical topological information in the form of persistence diagrams for real-world applications while maintaining the exponential speedup advantage of quantum computation; (3) proposing a novel hybrid paradigm of "classical precision guiding quantum efficiency." This method provides a feasible pathway for the practical implementation of quantum topological data analysis.

quant-ph↗

arXiv2025

Efficient Graph Optimization via Distance-Aware Graph Representation Learning

We propose an efficient framework that integrates distance-aware multi-hop message passing with dynamic topology refinement. Unlike standard GNNs that rely on shallow, fixed-hop aggregation, DRTR leverages both static preprocessing and dynamic resampling to capture deeper structural dependencies. A \emph{Distance Recomputator} prunes semantically weak edges using adaptive attention, while a \emph{Topology Reconstructor} establishes latent connections among distant but relevant nodes. This joint mechanism enables more expressive and robust graph representation optimization across evolving graph structures. Extensive experiments demonstrate that DRTR outperforms baseline GNNs in both accuracy and scalability, with at most 20\% computational overhead, especially in complex and noisy graph environments.

cs.LG↗

arXiv2025

LLMEasyQuant: Scalable Quantization for Parallel and Distributed LLM Inference

As large language models (LLMs) grow in size and deployment scale, quantization has become an essential technique for reducing memory footprint and improving inference efficiency. However, existing quantization toolkits often lack transparency, flexibility, and system-level scalability across GPUs and distributed environments. We present \textbf{LLMEasyQuant}, a modular, system-aware quantization framework designed for efficient, low-bit inference of LLMs on single-node multi-GPU, multi-node, and edge hardware. LLMEasyQuant supports a wide range of quantization methods -- including Symmetric Quantization, ZeroQuant, SmoothQuant, and SimQuant -- with unified interfaces for per-layer calibration, bitwidth assignment, and runtime adaptation. It integrates fused CUDA kernels with NCCL-based distributed synchronization and supports both static and online quantization. Empirical results show that LLMEasyQuant can achieve substantial speedup in GEMM execution, HBM load time, and near-linear multi-GPU scaling. Ablation studies further validate its ability to balance latency, memory, and accuracy under diverse deployment conditions. LLMEasyQuant offers a practical quantization serving system for scalable, hardware-optimized LLM inference.

cs.LG↗

arXiv2025

Observational properties and quasinormal Modes of the Hayward black Hole surrounded by a cloud of strings

In this work, we explored the Hayward black hole surrounded by a cloud of strings, with a focus on the effects of the regularization parameter $l$ and the string cloud parameter $a$ on its observational properties and quasinormal modes (QNMs). Utilizing the spacetime metric and geodesic equations, we calculated several geometric quantities characterizing the black hole. To visualize the observational appearance of the accretion disk, we employed the Novikov--Thorne model to simulate both its primary and secondary images. Furthermore, we analyzed the QNMs of the black hole under scalar and electromagnetic perturbations for different parameter values. The results indicate that as the regularization parameter $l$ increases, the outer horizon radius $r_{+}$, photon-sphere radius $r_{\text{ph}}$, critical impact parameter $b_{c}$, and innermost stable circular orbit $r_{\text{isco}}$ exhibit a gradual decrease, while the inner horizon radius $r_{-}$ and the real part of the QNMs frequency $ω_{r}$ increase. In contrast, as the string cloud parameter $a$ increases, $r_{+}$, $r_{\text{ph}}$, $b_{c}$, and $r_{\text{ isco}}$ demonstrate a rapid increase, whereas $r_{-}$ and $ω_{r}$ decrease. In both cases, the absolute value of the imaginary part of the QNMs frequency decreases with the increase $l$ or $a$. This work offers a theoretical foundation for understanding the coupling between regular black holes and surrounding string clouds.

gr-qc↗

arXiv2025

Path-Optimized Fast Quasi-Adiabatic Driving in Coupled Elastic Waveguides

Fast quasi-adiabatic driving (FAQUAD) is a central technique in shortcuts to adiabaticity (STA), enabling accelerated adiabatic evolution by optimizing the rate of change of a single control parameter. However, many realistic systems are governed by multiple coupled parameters, where the adiabatic condition depends not only on the local rate of change but also on the path through parameter space. Here, we introduce an enhanced FAQUAD framework that incorporates path optimization in addition to conventional velocity optimization, extending STA control to two-dimensional parameter spaces. We implement this concept in a coupled elastic-waveguide system, where the synthetic parameters-detuning and coupling-are controlled by the thicknesses of the waveguides and connecting bridges. Using scanning laser Doppler vibrometry, we directly map the flexural-wave field and observe adiabatic energy transfer along the optimized path in parameter space. This elastic-wave platform provides a versatile classical analogue for exploring multidimensional adiabatic control, demonstrating efficient and compact implementation of shortcut-to-adiabaticity protocols.

physics.app-ph↗

arXiv2025

PredicateFix: Repairing Static Analysis Alerts with Bridging Predicates

Fixing static analysis alerts in source code with Large Language Models (LLMs) is becoming increasingly popular. However, LLMs often hallucinate and perform poorly for complex and less common alerts. Retrieval-augmented generation (RAG) aims to solve this problem by providing the model with a relevant example, but existing approaches face the challenge of unsatisfactory quality of such examples. To address this challenge, we utilize the predicates in the analysis rule, which serve as a bridge between the alert and relevant code snippets within a clean code corpus, called key examples. Based on this insight, we propose an algorithm to retrieve key examples for an alert automatically, and build PredicateFix as a RAG pipeline to fix alerts from two static code analyzers: CodeQL and GoInsight. Evaluation with multiple LLMs shows that PredicateFix increases the number of correct repairs by 27.1% ~ 69.3%, significantly outperforming other baseline RAG approaches.

cs.SE↗

arXiv2025

Cognitive Load Traces as Symbolic and Visual Accounts of Deep Model Cognition

We propose \textbf{Cognitive Load Traces} (CLTs) as a mid-level interpretability framework for deep models, inspired by Cognitive Load Theory in human cognition. CLTs are defined as symbolic, temporally varying functions that quantify model-internal resource allocation. Formally, we represent CLTs as a three-component stochastic process $(\mathrm{IL}_t, \mathrm{EL}_t, \mathrm{GL}_t)$, corresponding to \emph{Intrinsic}, \emph{Extraneous}, and \emph{Germane} load. Each component is instantiated through measurable proxies such as attention entropy, KV-cache miss ratio, representation dispersion, and decoding stability. We propose both symbolic formulations and visualization methods (load curves, simplex diagrams) that enable interpretable analysis of reasoning dynamics. Experiments on reasoning and planning benchmarks show that CLTs predict error-onset, reveal cognitive strategies, and enable load-guided interventions that improve reasoning efficiency by 15-30\% while maintaining accuracy.

cs.AI↗

arXiv2025

SecuLEx: a Secure Limit Exchange Market for Dynamic Operating Envelopes

Distributed energy resources (DERs) are transforming power networks, challenging traditional operational methods, and requiring new coordination mechanisms. To address this challenge, this paper introduces SecuLEx (Secure Limit Exchange), a new market-based paradigm to allocate power injection and withdrawal limits that guarantee network security during time periods, called dynamic operating envelopes (DOEs). Under this paradigm, distribution system operators (DSOs) assign initial DOEs to customers. These limits can be exchanged afterward through a market, allowing customers to reallocate them according to their needs while ensuring network operational constraints. We formalize SecuLEx and illustrate DOE allocation and market exchanges on a small-scale low-voltage (LV) network, demonstrating that both procedures are computationally tractable. In this example, SecuLEx reduces renewable curtailment and improves grid utilization and social welfare compared to traditional approaches.

eess.SY↗

arXiv2025

Exposing Hallucinations To Suppress Them: VLMs Representation Editing With Generative Anchors

Multimodal large language models (MLLMs) have achieved remarkable success across diverse vision-language tasks, yet they remain highly susceptible to hallucinations, producing content that is fluent but inconsistent with visual evidence. Such hallucinations, spanning objects, attributes, and relations, persist even in larger models, while existing mitigation approaches often require additional finetuning, handcrafted priors, or trade-offs that compromise informativeness and scalability. To address this limitation, we propose a training-free, self-supervised method for hallucination mitigation. Our approach introduces a novel hallucination amplification mechanism: a caption is projected into the visual space via a text-to-image model to reveal implicit hallucination signals, serving as a negative anchor, while the original image provides a positive anchor. Leveraging these dual anchors, we edit decoder hidden states by pulling representations toward faithful semantics and pushing them away from hallucination directions. This correction requires no human priors or additional training costs, ensuring both effectiveness and efficiency. Extensive experiments across multiple benchmarks show that our method significantly reduces hallucinations at the object, attribute, and relation levels while largely preserving recall and caption richness, e.g., achieving a hallucination reduction by over 5% using LLaVA-v1.5-7B on CHAIR. Furthermore, results on diverse architectures, including LLaVA-NEXT-7B, Cambrian-8B, and InstructBLIP-7B, validate strong cross-architecture generalization. More importantly, when applied to hallucination-free captions, our method introduces almost no side effects, underscoring its robustness and practical plug-and-play applicability. The implementation will be publicly available.

cs.CV↗

arXiv2025

HSGM: Hierarchical Segment-Graph Memory for Scalable Long-Text Semantics

Semantic parsing of long documents remains challenging due to quadratic growth in pairwise composition and memory requirements. We introduce \textbf{Hierarchical Segment-Graph Memory (HSGM)}, a novel framework that decomposes an input of length $N$ into $M$ meaningful segments, constructs \emph{Local Semantic Graphs} on each segment, and extracts compact \emph{summary nodes} to form a \emph{Global Graph Memory}. HSGM supports \emph{incremental updates} -- only newly arrived segments incur local graph construction and summary-node integration -- while \emph{Hierarchical Query Processing} locates relevant segments via top-$K$ retrieval over summary nodes and then performs fine-grained reasoning within their local graphs. Theoretically, HSGM reduces worst-case complexity from $O(N^2)$ to $O\!\left(N\,k + (N/k)^2\right)$, with segment size $k \ll N$, and we derive Frobenius-norm bounds on the approximation error introduced by node summarization and sparsification thresholds. Empirically, on three benchmarks -- long-document AMR parsing, segment-level semantic role labeling (OntoNotes), and legal event extraction -- HSGM achieves \emph{2--4$\times$ inference speedup}, \emph{$>60\%$ reduction} in peak memory, and \emph{$\ge 95\%$} of baseline accuracy. Our approach unlocks scalable, accurate semantic modeling for ultra-long texts, enabling real-time and resource-constrained NLP applications.

cs.AI↗

arXiv2025

Scientific Objectives of the Xue-shan-mu-chang 15-meter Submillimeter Telescope

Submillimeter astronomy is poised to revolutionize our understanding of the Universe by revealing cosmic phenomena hidden from optical and near-infrared observations, particularly those associated with interstellar dust, molecular gas, and star formation. The Xue-shan-mu-chang 15-meter submillimeter telescope (XSMT-15m), to be constructed at a premier high-altitude site (4813 m) in Qinghai, China, marks a major milestone for Chinese astronomy, establishing the China mainland's first independently developed, world-class submillimeter facility. Equipped with state-of-the-art instruments, XSMT-15m will address a diverse range of frontier scientific questions spanning extragalactic astronomy, Galactic structure, time-domain astrophysics, and astrochemistry. In synergy with current and forthcoming observatories, XSMT-15m will illuminate the formation and evolution of galaxies, unravel the physical and chemical processes shaping the interstellar medium, and explore transient phenomena in the submillimeter regime. These capabilities will advance our understanding across extragalactic astronomy, Galactic ecology, astrochemistry, and time-domain astrophysics, inaugurating a new era for submillimeter research in China and the northern hemisphere.

astro-ph.GA↗

arXiv2025

Emphasising Structured Information: Integrating Abstract Meaning Representation into LLMs for Enhanced Open-Domain Dialogue Evaluation

Automatic open-domain dialogue evaluation has attracted increasing attention, yet remains challenging due to the complexity of assessing response appropriateness. Traditional evaluation metrics, typically trained with true positive and randomly selected negative responses, tend to assign higher scores to responses that share greater content similarity with contexts. However, adversarial negative responses, despite possessing high lexical overlap with contexts, can be semantically incongruous. Consequently, existing metrics struggle to effectively evaluate such responses, resulting in low correlations with human judgments. While recent studies have demonstrated the effectiveness of Large Language Models (LLMs) for open-domain dialogue evaluation, they still face challenges in handling adversarial negative examples. We propose a novel evaluation framework that integrates Abstract Meaning Representation (AMR) enhanced domain-specific language models (SLMs) with LLMs. Our SLMs explicitly incorporate AMR graph information through a gating mechanism for enhanced semantic representation learning, while both SLM predictions and AMR knowledge are integrated into LLM prompts for robust evaluation. Extensive experiments on open-domain dialogue evaluation tasks demonstrate the superiority of our method compared to state-of-the-art baselines. Our comprehensive ablation studies reveal that AMR graph information contributes substantially more to performance improvements. Our framework achieves strong correlations with human judgments across multiple datasets, establishing a new benchmark for dialogue evaluation. Our code and data are publicly available.

cs.CL↗

arXiv2025

Crafting Customisable Characters with LLMs: A Persona-Driven Role-Playing Agent Framework

Large Language Models (LLMs) demonstrate remarkable ability to comprehend instructions and generate human-like text, enabling sophisticated agent simulation beyond basic behavior replication. However, the potential for creating freely customisable characters remains underexplored. We introduce the Customisable Conversation Agent Framework, which employs LLMs to simulate real-world characters through personalised characteristic feature injection, enabling diverse character creation according to user preferences. We propose the SimsConv dataset, comprising 68 customised characters and 13,971 multi-turn role-playing dialogues across 1,360 real-world scenes. Characters are initially customised using pre-defined elements (career, aspiration, traits, skills), then expanded through personal and social profiles. Building on this, we present SimsChat, a freely customisable role-playing agent incorporating various realistic settings and topic-specified character interactions. Experimental results on both SimsConv and WikiRoleEval datasets demonstrate SimsChat's superior performance in maintaining character consistency, knowledge accuracy, and appropriate question rejection compared to existing models. Our framework provides valuable insights for developing more accurate and customisable human simulacra. Our data and code are publicly available at https://github.com/Bernard-Yang/SimsChat.

cs.CL↗